Each depositor insured to at least $250,000 per insured bank

Because my father spent his career in the Air Force, I have had some interest in military history -- in the strategies for winning and the strategies that resulted in defeat. In military history, the First World War stands out because of the extraordinary discrepancy that existed between the conditions of warfare on the Western Front and the strategies and tactics that dominated the minds of military decision makers. For the first three years of the war, thinking on both sides emphasized attack -- an artillery barrage, followed by the infantry moving in a headlong assault with bayonets, followed by the cavalry. Artillery, bayonets, cavalry -- the ABCs of warfare -- what could be simpler? The problem was that -- by 1914 -- barbed wire, machine guns and trenches had made this approach obsolete. Officers, however, continued to believe that such an attack would be successful if the morale of the attackers could be brought to a sufficiently high pitch -- after all, it had worked for Napoleon. Ideas have consequences -- and armies on both sides suffered the largest military casualties in history as a result of this approach. Despite millions of casualties, the front -- extending from the English Channel on one extreme to Switzerland on the other -- remained almost unchanged for more than three years.

Neither military strategy nor bank supervision exists in a vacuum. To be effective, both need to be shaped to current conditions in the field. If we as bank supervisors have learned anything in the last 15 years, it is that being out of touch with current realities endangers the banking system and our own credibility.

No one has repealed the business cycle -- we will continue to experience good times and tough times in the economy. Memories are short, and times are never so good that banks face no problems -- incipient, developing, or full blown. The critical issue in bank supervision is steering a moderate course between the theoretical extreme where there are no problems and the extreme where everything is a problem.

A moderate course is appropriate -- a course that recognizes that when things are going badly, the pendulum has a way of swinging back -- a course that recognizes that when things are going well, the pendulum will swing the other way, too. The moderate course is the course justified by critical analysis and sound judgment. We bank regulators have one of the few jobs in the world where we can be described as flat-footed financial Keystone Kops one year and "regulators from hell" the next. We avoid that kind of labeling only by always being realists -- always being somewhat skeptical of conventional wisdom and never being wholly optimistic nor solely pessimistic. How do we steer such a moderate course?

First -- in the good times like those we are experiencing now -- we need to develop our expertise and expand our knowledge. We need to benchmark where banking is now -- and where it has been previously -- to give us perspective on where it is going in the future.

At the Federal Deposit Insurance Corporation, we are addressing this need in a number of ways.

For example, we are enhancing our ability to leverage existing statistical and analytical resources, both within and outside the FDIC. All our agencies have generated a treasury of data on the banking and thrift industries, including a data base that the FDIC is developing on the causes of the large number of failing and failed institutions in the 1980s and early 1990s. On an ongoing basis, we are surveying our examiners on credit underwriting practices and standards. We have created a Division of Insurance to analyze risks to the insurance funds from a more comprehensive perspective than we have done to date. This new division will generate data and gather and synthesize analysis from outside the FDIC -- information generated by analysts such as yourselves. Our goal is to "bridge the gap" that currently separates the "macro" perspective of economics and market trends from the "micro" perspective of bank examinations in ways that will translate information into guidance that examiners can use in assessing and monitoring risks and evaluating internal controls in institutions with different levels and types of risk exposure. Put simply: We want the different parts of our agency talking -- and working -- together.

Further, each of the agencies collects and analyzes data from different perspectives. Regular efforts to compare trends and pass those on to the three other supervisory agencies would be helpful to all of us.

That brings me to the second element in steering a moderate course during good times -- being alert to problems and doing something about them before they result in damage to banks and the banking system. In the late 1970s and early 1980s, examiners in the Midwest witnessed a dangerous credit practice -- agricultural banks were lending without requiring well-defined repayment programs and based upon inflated land values. Fueled by export growth and rising commodity prices, U.S.agriculture in the 1970s enjoyed a boom, which in turn caused the price of farm land to rise significantly and prompted many farmers to borrow to expand operations using inflated real estate values to support increases in debt. The boom ended, land values declined, and cash flow was insufficient to repay the debt. Agricultural lenders, in turn, experienced large loan losses and agricultural banks accounted for 32 percent of bank failures in 1984, 54 percent in 1985, 41 percent in 1986 and 30 percent in 1987. Substitute the words "oil boom' for "agricultural boom" and you have a similar story -- from 1980 through 1989, 535 banks failed in Texas, Oklahoma, and Louisiana -- half of all U.S. bank failures during the period. In Texas alone, 349 banks failed and an additional 79 required FDIC financial assistance. Some of the failures were of agricultural banks, but the majority succumbed to problems related to energy. Substitute the words "real estate boom" for "oil boom" and we begin moving from the late 1980s into the nineties. From 1980 to 1990, real estate loans at banks rose from 14.5 percent of assets to 24.5 percent of assets, with a shift away from home mortgage lending to more volatile construction and commercial lending, accompanied by a relaxation in underwriting standards for construction and commercial real estate.

All of this, of course, is an old story, and one burned into the memories of many of us here today. I, too, have personal memories of the crisis that visited U.S. banking in August, 1982, when Mexico announced that it would be unable to meet its principal payments to foreign creditors. That came after the banks were actively encouraged by some U.S. policymakers in the 1970s to recycle petrodollars. The banks unfortunately concluded that the fastest way to lend dollars was through balance of payments financing without any clear source of repayment -- and with too much faith in optimistic expectations about the economic and financial prospects of many developing countries. By 1982, the non-trade exposure of the average U.S. money-center bank to non-OPEC developing countries was 227 percent of equity capital and reserves.

In the years 1980 through 1995 -- 1,626 Bank Insurance Fund-insured institutions failed or received assistance transactions. They held more than $304 billion in assets. Of these institutions, 624 -- holding more than $165 billion in assets -- were national banks. More than 927 of the institutions -- holding more than $58 billion in assets -- were state-chartered commercial banks. Savings banks -- state and federal -- made up the remainder. The size of the average failure was about $150 million in assets, but that average masks two extremes. Of the failed institutions, 1,307 had assets of less than $100 million, together holding $36 billion in assets. The 42 institutions that exceeded $1 billion in assets together held $193 billion in assets. These numbers do not include the many banks -- some quite large -- that had near-death experiences and survived only because some of you in this room helped nurse them back to health.

My point in revisiting the past is a simple one: in agriculture, oil, real estate and developing country lending, inflated values and expectations provided a false sense of security. In each case, we thought the crisis of the moment was special -- there was a tendency not to draw from each experience the broader implications about trends in bank lending -- of not noticing -- until it was too late -- the general phenomenon of overextended lending and weak capital. By the time those of us working on these issues realized the individual crises were linked by common forces that were pushing banks to exploit new areas of lending beyond sustainable levels, the problems had become cumulative.

Neither we nor the industry we supervise can afford being so wrong again. The speed of technology and the rapid innovations in the marketplace mean that trouble could come quickly and in large numbers. We need to avoid being that wrong again by monitoring trends more broadly and taking specific action on the information we receive.

Several weeks ago, I was talking with a banker in Texas -- his institution, approximately $200 million in assets -- is within an hour's drive of Dallas/Ft. Worth. I asked him how he made it through the energy/real estate storm that blew through Texas, and he replied: "The key thing was that, when things looked so good, that's exactly when you should be worried. I was involved with the state bankers association, so I heard stories of things looking worse in certain parts of Texas -- a comment here, a comment there. I took stock to see what was happening elsewhere and to consider its relevance to us. I knew that we were not smart enough or lucky enough to keep it from happening to us -- if other folks were having problems, we would, too. I saw problems elsewhere in real estate, and lowered our exposure -- we hid behind a log for a while. If you wait until the cows are out of the barn, it's too late. That's what helped us through the eighties."

This banker survived by being alert to problems and by taking preemptive action before problems grew. He survived without our help. I believe others might have survived with our help if they had been encouraged to take a broader perspective. Our efforts to identify, measure and monitor risks in and to the banking system will provide a basis for notices to banks on trends that may affect the way they do business so that they can respond to changing circumstances before problems arise. Such notices can be purely economic -- the results of our modeling on bank failures and other economic trends, analysis of connections like rolling recessions of the kind in the 1980s and 1990s -- but they can also discuss the effects of other types of events -- legislation, for example. Early and thorough analysis of the difficulties thrifts would encounter as a result of the timing of the elimination of interest rate restrictions -- and of the effects of the sudden shift in the real estate investment climate brought about by the Tax Reform Act of 1986 -- might have avoided at least some of the trouble in the banking system over the last fifteen years.

The third element in steering a moderate course in banking supervision is to be just as realistic when the cycle turns down as we are when the cycle is on the upswing -- that is to say, while being alert to problems, we should not fall into the mindset that problems lurk under every rock and in every loan file -- not all technical violations of statute, regulations or examination guidelines present safety and soundness problems. One of my predecessors -- Bill Taylor -- used to say when finishing an examination where the bank received the highest supervisory rating: "We didn't find the problems this time, but we'll be back." That approach may have made sense during a time of crisis, but conditions now do not justify that approach.

Comptroller of the Currency Gene Ludwig illustrated that mindset in a speech a number of years ago when he noted that one of the OCC's examiners had criticized a bank for not having a branch policy -- which was perfectly understandable under the circumstances because the bank had no branches. A predisposition to find fault -- a confrontational approach -- where it is not warranted is just as incorrect as ignoring problems that are there. It may also be just as destructive because it may cause a negative reaction that could leave our safety and soundness policies impaired by a legislative reaction.

At the FDIC, we have emphasized the need for clear communication with bankers and the need to project a helpful attitude when corrections are necessary. We also try to make sure that minor matters do not get overemphasized. Like the rest of you, we have increasingly emphasized the importance of risk-focused examinations and of avoiding a checklist approach to examinations.

Many people have perfect hindsight -- or at least claim to -- but no one has perfect foresight. The agencies here must work together to assure that we can better draw the distinction between what we see as problems and what we do not see as problems. In this exercise, different points of view enhance perspective, but we need to make sure we come together after all the issues are resolved to assure consistency in banking supervision.

The fourth element in steering a moderate course is to stick to basic principles of bank supervision both in good times and in tough times. One such principle is that every bank needs to be examined --on-site -- routinely and regularly. It is a mistake to assume that smaller institutions are inherently less in need of regular and routine on-site examinations simply because they are small and not as complex. As I said earlier, 1,307 of the 1,626 BIF-insured institutions that failed from 1980 through 1995 had assets of less than $100 million. Small banks deserve our supervision and guidance as much as large banks do. I have talked with a number of bankers in smaller institutions who tell me that they become uncomfortable if too much time lapses between examinations.

Another basic principle of bank supervision is that care should be given in chartering new banks. From 1980 through 1989, there were 673 new commercial banks chartered in Texas -- 511 national banks and 162 state-chartered banks -- about a quarter of all new banks chartered in that decade. Of those new Texas charters, 142 -- close to one out of five -- failed by 1990. Certainly, we cannot eliminate bank failures entirely -- any attempt to do so would starve the economy of bank credit -- but we can protect the banking system from unnecessary failures -- failures from problems that could have been avoided or addressed before they threatened the viability of the institution.

Soon after Bill Taylor became Chairman of the FDIC, he visited our New York regional office in connection with the resolution of a savings institution. Bill -- being Bill -- wanted to check out some of the branch offices of the thrift to see how business was going. He told Nick Ketcha -- then head of our New York region -- to call for the car and driver to take them around.

"Bill," Nick replied, "remember, this is the FDIC -- we don't have a car and driver."

They took the subway instead.

When I visited our New York office for the first time, Nick Ketcha also took me on the subway.

We may travel by varying paths, but as bank supervisors we all should be going in the same direction. The accidents of history created four different federal supervisors for two closely related species of financial institutions. This situation gives us the opportunity for inclusive diversity -- a chance to share strengths. It also gives us the opportunity to go separate ways.

Never has it been more important for us to work together than today. Never has it been more critical for us to set one consistent, sound, appropriate, reasonable and moderate course for banking supervision than today.

We have to send clear signals to bankers and others who are trying to accommodate their behavior to our direction. Overlapping or conflicting signals will, at best, confuse them and, at worst, may lead to behavior we do not want to see.

During the Crimean War, an event that was to be immortalized in "The Charge of the Light Brigade" began with a confused directive from officer to officer to "go for the guns." The order was to the cavalry to retrieve a number of British cannon aimed at Russian lines. Instead, the cavalry officers took the order to mean that they should charge Russian artillery positions. The result was a military disaster still remembered 150 years later.

Consistency results in clarity. When so many have so much riding on us, we must be clear.

It is with great pleasure, therefore, that I welcome all of you to this interagency meeting -- it has been several years since we have come together in this way. I hope that this meeting marks the resumption of annual meetings. Hosting this meeting clearly emphasizes the importance that we at the FDIC place on interagency cooperation and coordination.

Our discussions over the next two days will range from designing a new breed of CAMEL to supervising banking in cyberspace -- from the lessons of Daiwa to de novo applications issues. The involvement of principals from all of the agencies emphasizes the strong desire for consistency.

We are here today because we share a calling. It is not one that receives great public applause, and it is not one for which we receive great material reward. Our greatest compensation is most often the respect and understanding of our peers. We know that without the stability that we bring to an inconstant world, Americans would suffer -- just as they suffered before our regulatory system was created -- families often losing their homes and children often losing the opportunity that higher education provides. At the end of the day, we have the satisfaction of knowing that what we do is important, that we make a difference in people's lives. Government exists to serve the public and what joins us all together is that we take that service seriously.